Detection and Continual Learning of Novel Face Presentation Attacks
This addresses the problem of adapting face antispoofing systems to new attacks for security applications, but it is incremental as it builds on existing continual learning and anomaly detection methods.
The paper tackles the vulnerability of face presentation attack detection systems to novel, unseen attacks and their inability to adapt post-training, by enabling a deep neural network to detect anomalies as potential new attacks and using experience replay to update the model without forgetting past attacks. Experimental results show effectiveness on two benchmark datasets and a new dataset with varied attack types.
Advances in deep learning, combined with availability of large datasets, have led to impressive improvements in face presentation attack detection research. However, state-of-the-art face antispoofing systems are still vulnerable to novel types of attacks that are never seen during training. Moreover, even if such attacks are correctly detected, these systems lack the ability to adapt to newly encountered attacks. The post-training ability of continually detecting new types of attacks and self-adaptation to identify these attack types, after the initial detection phase, is highly appealing. In this paper, we enable a deep neural network to detect anomalies in the observed input data points as potential new types of attacks by suppressing the confidence-level of the network outside the training samples' distribution. We then use experience replay to update the model to incorporate knowledge about new types of attacks without forgetting the past learned attack types. Experimental results are provided to demonstrate the effectiveness of the proposed method on two benchmark datasets as well as a newly introduced dataset which exhibits a large variety of attack types.